Bug Reports

Thanks, Phungus (religion-popping issue). I found a forum page followed by Afforess and put a comment there, not knowing how to PM him. Since you've sent it to Glider, we'll see what can be done.

A related nit to changing how religions pop is the fact that the AI logic will still occasionally have them buy meditation or polytheism (or code of laws, etc., in a later era), if they're spiritual and really want a religion. That won't work anymore with "buying tech" not getting you a religion, so AI logic should no longer have them do this. (In one game, I never saw Buddhism, tho I was first human to research it, so I'm pretty sure AI bought meditation and thus turned it off for good -- or at least it's off until the other 6 all get invented [we didn't make it to Islam to see if Budd popped somewhere at that point].)
 
I have a game crash issue.

The attached save crashes after ending the turn when using 0.9.9b or 0.9.9c.
 

Attachments

  • LoR-1-Carl AD-1735.CivBeyondSwordSave
    546.5 KB · Views: 236
I have a game crash issue.

The attached save crashes after ending the turn when using 0.9.9b or 0.9.9c.

It definitely crashes. When debugging, the crash happens in the disassembler, and seems to be deep in the .exe. Whenever this happened before it was always an art issue, but checking the logs, I came up with nothing (every other crash like this I was able to find the bad art caused by a unit/building in the resmgr.log, or the LSystem.log). I have no idea where to proceed to hunt this down, as there is nothing in the logs that show anything is wrong, or point to any bad art, and the debugger crash only shows me unreadable machine code, since the crash seems to be happening in the .exe which we don't have exposed source code for.

Any ideas?
 
Your "nothing in the logs" is also what I saw.

The only hint is maybe what is in the PythonDbg.log file since at the end of that it looked like it had produced a few messages while processing the turn (Washington grew in pop and a few other lines of stuff after that - I don't have a log for it at the moment, but I think it had moved on to at least one more player based on a city name in there that wasn't mine). Whatever happened happened after the last thing in there but before the turn processing completed.

In the debugger, if you can see things in the call stack before the final EXE call frame (and you should be able to), particularly things that are in the DLL, then you should be able to tell what sort of thing it was doing. You should also be able to move up to that code and check which player it was and what building/unit/whatever it was dealing with at the time. Unless the crash resulted in (or from) a corrupted call stack, or something along those lines.

Or you may have to resort to putting a breakpoint in and stepping through the turn processing code at a high level, like CvGame::doTurn. The per-player loop should let you know which player's turn processing is having the problem (the one it is working on when it fails). Then the next time through you can try stepping down into the problem player's turn processing in CvPlayer:setTurnActive and doTurn to see more of what it was doing at the time.
 
I've only ever had luck stepping through code. I can't even find the call stack... I think I may be debugging improperly, but haven't really needed to get very in depth as I've only worked with my own code, or bits of BtS code that interacts with my code.

The only thing in RevDCM (and thus LoR), that I don't fully trust is the WoC code, and the ethnic city art. This crash isn't caused by ethnic city art. I think what I may need to do is just roll up my sleeves, and extricate WoC from LoR. I also think this may fix a nagging MP issue when files are in the custom assets. The problem with doing this though is that it's alot of work, and will make keeping LoR up to date with RevDCM much more difficult, as there would significant deviations in the code base.

Not sure, I'll have to think on things. For the most part though, I'm stumped on this crash, as there is nothing but random machine code to work with. Every other crash I"ve had reported, since back when I started with WolfRevolution, there was at least some starting point.
 
On Deity, the start as minor tribes option appears to be taking away the 4 free techs the AI is supposed to start with as well as taking away their scouts.

Ran into an odd bug with Revolutions. Had a low culture city that went into revolt as it was next to a vassalized high culture city. City revolted, options listed included
1) Grant city independence
2) Refuse and city will rise up against you as a new civ
3) Declare war on the civ whos culture is causing the problem (in this case my vassal)

I picked going war with my vassal planning to finish him off and remove the problem. I thought it would cancel the vassal agreement. It did not. Instead a bunch of enemy units from a competing civ based half way around the world (Who I was also at war with spawned).

Civilopedia entry for Egypt's war Chariot says it is immune to first strikes but actual unit bonus is +50% vs archery units.
 
WoC definitely comes with a few bugs (including some nasty MP ones that took me forever to track down), but I'm pretty sure the WoC code isn't active during actual gameplay, just loading the game and saves, so it's not going to cause CTD's.

Well, unless WoC does something funny with the art loading - I know it did. I tried to fix in RevDCM the best I could - but you may want to do a diff on the WoC files between AND and RevDCM.
 
OK, I have successfully extirpated WOC, and updated to the 2.9 RevDCM core. I can't be sure if this fixes the crash reported above because it breaks save game compatibility, but I'm pretty sure it will; since WOC code is the only code I don't really trust. I guess there is some stuff in Revolutions that could be suspect as well, so I will, over the next few months, work on rebuilding RevDCM in the DLL, instead of around python where it mostly lives now. But that is a long term project.

In regards to religion founding, I have rebuilt it again, and I believe it should work now in multiplayer, and fix the bugs reported above regarding religion founding in multiplayer games.

Because I have made some drastic changes to the code base (removing WOC), I'm not fully confident in the stability of this version. I also lack the time to properly play test it, as my computer is very, very slow. So I'll hold off on making this version the "official" release. Please feel free to playtest, and let me know any bugs you come across or balance issues:

http://www.filefront.com/17914869/Legends-of-Revolution-v0.9.10-Setup.exe/
 
On Deity, the start as minor tribes option appears to be taking away the 4 free techs the AI is supposed to start with as well as taking away their scouts.

This is intended, otherwise at Emperor+ with start as minors on the AI simply b-lines your starting city with archers immediately, and you die.

Ran into an odd bug with Revolutions. Had a low culture city that went into revolt as it was next to a vassalized high culture city. City revolted, options listed included
1) Grant city independence
2) Refuse and city will rise up against you as a new civ
3) Declare war on the civ whos culture is causing the problem (in this case my vassal)

I picked going war with my vassal planning to finish him off and remove the problem. I thought it would cancel the vassal agreement. It did not. Instead a bunch of enemy units from a competing civ based half way around the world (Who I was also at war with spawned).
Hmmm, without seeing the save I'm not sure, but this may be intended behavior, since you can't attack your own vassals (if it wasn't a vassal, a war would have been launched, as I've seen this before). Anyway I plan on rebuilding the Revolutions code, so I'll keep this in mind when doing so.

Thanks for the reports.
 
Anyone found any issues with the 1.0 dev build? If it is working fine, I'll go ahead and set that to the main download, but just want to make sure no problems have been found in it.
 
Reveloutions don't seem to be working for me. I got this message and duly refused the Deep South attempting to secede from the already seceding South (redundancy much?) I expected them to break out into reveloution a few turn later as starvation worsened, the war dragged on with heavy unit losses (though victorious) I mean the population of those cities dropped down to 1 but that was it. They never revolted. Also in the event log it says something like "Barbarian Civ error 1023" or something along those lines. So I assume that the reveloution event is trying to fire but it's not working. Any idea why? Is it because this is a scenario? I mean I've already installed this mod twice.



Uploaded with ImageShack.us
 
Is that from the 1.0 development build, or the current "official release"? If it's the Dev build, I suppose something could be wrong. If it's not, then the issue is one of revolution type, and teh revolution just never gets enough "oomph" to get pushed over into a full revolt.
 
Hello.

Something weird, that happened to me repeatedly.

Look at the "before" Screenshot.

To help the newly founded city of Grenoble (far left), I built 2 workboats in Marseilles (far right) and sent them to Grenoble. As you can see, the first one has already arrived and is already working a clam. The second one is on its way (lower center), and as you can see Grenoble is a valid destination.

But before that workboat completes its travel, I discover OPTICS. And now, the destination is "forbidden", as can be seen in the "after" screenshot. :crazyeye:
I can't imagine this being WAD.

I also have a savegame just one turn prior to Optics. I'll upload it if requested.
 

Attachments

  • optics bug  before.jpg
    optics bug before.jpg
    72.7 KB · Views: 244
  • Optics bug After.jpg
    Optics bug After.jpg
    76.5 KB · Views: 338
In my current game, v0.9.10, I spotted two things:

1) Great wall gives 200% more (not 100% more) experience to GG bar for the battles in my cultural borders. Battles outside the borders give the same amount of XP to unit and to the GG bar. I never really paid attention to this, so I am not sure if this is new or came with this build. (Leader is Charlemagne, philosophical and aggressive)

2) Tech deals with the Ruthless AI on. I got one deal (Writing partially and Archery) before I was able to research Alphabet. Since then I am unable to get any tech deal from AI, all techs are "we would have nothing to gain". Even though I had or still have something nice too. I googled what the ruthless AI really is and found Aforess explanation for AND, where he says it should be more about war, but the AI would want actually MORE tech trading.

Otherwise, everything seems ok so far with this build, revolts do happen for the other civs and for me will probably strike soon too :)

will provide save if needed, when I get to my desktop computer.
 
The harbour seems to give me negative amounts of food and the factory seems to give me positive amounts of food (city is unhealthy): :confused:


Uploaded with ImageShack.us


Uploaded with ImageShack.us

You're reading it wrong. The Harbor is providing against your LACK of food. You're negative because of accumulated unhealth. The +2 Health you're receiving from the Harbor is also reducing your food deficit. The factory is doing the same, as its +1 Unhealth is adding to your food deficit.

Clear? :crazyeye: :D
 
Hi, I'm not sure if these are bugs or if these features were taken out in 0.9.9, but I could find no mention of them in the changelog. Is dynamic civ names still in this mod? The option to enable it doesn't appear anymore for me. I love this feature and want it back. :( Also, was Hitler taken out as a leader? (I'm just wondering, I'm not a Neonazi, I swear!)

I was also wondering if some of the custom graphics aren't supposed to work in the early part of the game? None of the German skins seem to work until later in the game. However, for America all of them do work except for the axeman and spearman. I can understand the civ having a more generic, primitive look in the ancient era, but for some American skins and no German skins to work seems unbalanced. I haven't played the other civs enough to see if anything is missing. No, I don't have frozen animations on.

I use Windows 7 64bit. I've uninstalled and reinstalled, so it's not a result of permission problems. Any ideas?
 
Hi, I'm not sure if these are bugs or if these features were taken out in 0.9.9, but I could find no mention of them in the changelog. Is dynamic civ names still in this mod? The option to enable it doesn't appear anymore for me. I love this feature and want it back. :(

I was also wondering if some of the custom graphics aren't supposed to work in the early part of the game? None of the German skins seem to work until later in the game. However, for America all of them do work except for the axeman and spearman. I can understand the civ having a more generic, primitive look in the ancient era, but for some American skins and no German skins to work seems unbalanced. I haven't played the other civs enough to see if anything is missing. No, I don't have frozen animations on.

I've uninstalled and reinstalled, so it's not a result of permission problems. Any ideas?

Dynamic Civ Names is in there; it has to be enabled in the options screen.

Can't tell you anything about the art, however. Screenies?
 
American archer, axeman, and spearman. The archer works from the beginning of the game but the spearman and axeman do not. (The swordsman also works.)



German archer, axeman, spearman, and swordsman in the classical period all with the old skins.



As soon as I hit medieval, their custom skins work and there's a weird glitch where you can see their skins running to the tile from off-screen (not pictured).



No dynamic civ names option.





And finally, no Hitler.



Also, when I was testing the German skins I noticed another bug that I thought I'd solved. Whenever I choose "start as minor civs", it randomizes the city names. Notice in the German screenshots how my capital is Bochum and my second city is Berlin. For some reason this didn't happen when I tested American, although it has in the past. The same thing happens to the AIs and to all civs.

What could be causing all of this? The only thing I've altered with this mod is replacing the title screen with the original Earth screen and replacing the music with "Baba Yetu", but these bugs have occurred immediately after reinstalling before I changed anything, so I don't think it could be that.

:confused: :confused: :confused: :confused: :confused:
 
American archer, axeman, and spearman. The archer works from the beginning of the game but the spearman and axeman do not. (The swordsman also works.)



German archer, axeman, spearman, and swordsman in the classical period all with the old skins.



As soon as I hit medieval, their custom skins work and there's a weird glitch where you can see their skins running to the tile from off-screen (not pictured).



No dynamic civ names option.





And finally, no Hitler.



Also, when I was testing the German skins I noticed another bug that I thought I'd solved. Whenever I choose "start as minor civs", it randomizes the city names. Notice in the German screenshots how my capital is Bochum and my second city is Berlin. For some reason this didn't happen when I tested American, although it has in the past. The same thing happens to the AIs and to all civs.

What could be causing all of this? The only thing I've altered with this mod is replacing the title screen with the original Earth screen and replacing the music with "Baba Yetu", but these bugs have occurred immediately after reinstalling before I changed anything, so I don't think it could be that.

:confused: :confused: :confused: :confused: :confused:

Not that options screen. The BUG Menu in game.

I've got no idea about the other stuff, as it all seems to work fine for me (I always play Start as Minors). What version are you using?
 
Top Bottom